1. Triangulating touchstones for critical criteria
If International Humanitarian Law (IHL) is to be morally grounded, as well as embodying de facto political compromises, the content of IHL needs to be as close to the content of morality as it can be.Footnote 1 But how close is that? Laws, and international laws more specifically, and international laws for the conduct of war most specifically, face their own inherent necessities in order to operate effectively in the particular political context in which they are expected to function. Some militarists will ask: is IHL not permissive enough in order to, for example, allow the defeat of terrorists? Some moralists will ask: is IHL too permissive from a moral point of view? If IHL allows actions that there is no moral justification for allowing – certainly if it allows very many such actions – IHL lacks moral grounding. And morality might then require decent people to refrain from all or some of what IHL permits. IHL permits killing, maiming, and destruction of property – only some killing, some maiming, and some destruction of property, of course, but much more of each kind of action than is allowed outside war. Is the content of IHL merely an arbitrary political compromise between the requirements of effective military action in an inter-state arena and the requirements of ordinary morality? Or can the substantive content of the political compromise itself somehow be grounded in underlying moral principles?Footnote 2
The thesis of this article is that it is impossible to decide whether the existing laws for the conduct of war – or even the best possible laws of war – are morally too permissive without having a grasp on at least two of the broad requirements that morally acceptable international laws of war must satisfy.Footnote 3 These requirements have at least three analytically separable but practically interacting sources: what is needed in order for them to be morally acceptable, what is needed in order for them to function as law in the international arena, and what is needed in order for them to function in the circumstances of war. None of these three touchstones can be ignored. In particular, one cannot, as a number of contemporary analytic philosophers seem to wish, simply specify what seems morally best, completely ignoring the undeniable facts that the crucial requirements must be able to function within combat, and must be embodied in laws governing inter-state behaviour. Each of these three sources of desiderata is obviously enormously complex, and we could not hope here to consider anything more than especially relevant selected general aspects of each of the three areas: morality, international politics, and war. The next section focuses on a minimum demand of morality, and the third, on a minimum demand of international politics. Then, I turn to a possible trade-off created by the tension between these two demands and, finally, examine its embodiment in the contemporary case of terrorism/anti-terrorism as a means of drawing together the demands of war, morality, and international politics in a concrete case.
2. Two essential criteria for IHL
2.1. The laws of war and morality ought to be as similar as possible, but will necessarily be radically different
IHL is providing rules for the conduct of war. If these rules are to be morally acceptable, it is natural to assume, on the one hand, that these rules should be as similar to the rules for the remainder of life as, in the circumstances of war, they can be. This rests on the fundamental prior assumption that ‘morality is all of a piece’Footnote 4 – we do not confront radically and irreducibly different moralities for different arenas of life. Of course rules must, on the other hand, always be appropriate for the circumstances in which they apply. Even outside warfare, circumstances differ, and the concrete specifications of rules (moral and legal) must differ accordingly; for example, rules for emergency circumstances may well differ significantly from rules for ordinary circumstances.Footnote 5
For example, a war is in effect one extended emergency – or rather a series of successive emergencies – and the fact that the circumstances regularly constitute an emergency is taken into account during the initial formulation of the basic rules. The rules for war are shaped from the beginning to be rules for emergencies, so that further exceptions will not need to be made. IHL rightly takes for granted that attempts to make judicious ad hoc exceptions amidst great danger would be prone to error. Consequently, as is widely recognized, most laws of war can be exceptionless – can prohibit the making of additional, supposedly ‘emergency’ exceptions to the content of the rule as specified.Footnote 6 ‘Never target civilians.’ ‘Always choose the military objective the attack on which will cause the least civilian loss.’ The exceptionless character of most central laws of war is an important respect in which laws of war are stricter than nearly all moral rules for ordinary circumstances other than war, which allow exceptions but do not already contain subsidiary provisions for the various kinds of exceptions.Footnote 7 Exceptions to rules for ordinary circumstances must be judged case by case. Exceptions, if any, are built into rules for war, because the circumstances of combat are obviously not a context for moral reflection. Therefore, laws of war are, on the one hand, rigid in form and in that respect stricter than moral rules for normal domestic situations.
On the other hand, the laws of war are for the most part more permissive in content, and in that respect less strict, than moral rules for ordinary circumstances. The laws of war presuppose the acceptability of resort to broad forms of violent self-help. The decisive reason why in domestic life one can normally be reasonably expected and morally required to refrain from violent forms of self-help is because the background institutional context, much of it structured by further layers of laws with no analogues in the international arena, generally provides help in the form of the relatively impartial civil institutions, like police and courts, that contribute to the rule of law.Footnote 8 Narrowly focused urgent self-defence is, of course, an emergency exception within the domestic context, and unlike most exceptions in that arena it has relatively well-worked-out rules, perhaps because the stakes for the rule of law and the general prohibition against self-help in domestic law are so high if this exception is not strictly contained.
War breaks out where impartial institutions are yet to be created or any existing impartial institutions are thought by at least one party to have failed to protect their vital interests, apparently leaving violent self-help as the only resort. Our question will be whether the way that the laws of war provide for self-help is too permissive to be morally justifiable even in their own context. But clearly if there are to be any rules at all for war, their content will be significantly different, precisely in that they allow for broad forms of violent self-help, from the rules (legal and moral) for most of the rest of life where one's life is lived within a peaceful domestic society structured by institutions that provide non-violent solutions to many conflicts, relatively strict enforcement against those who resort to violent self-help in any situation other than self-defence, and a reasonably high probability of punishment for convicted violators.Footnote 9
Wars consist, as the common saying goes, of ‘killing people and breaking things’. Such widespread violence is firmly prohibited by the moral rules for ordinary circumstances. The divergences between IHL and the rules for peaceful domestic life are striking.Footnote 10 One constraint on the conduct of civil life is a fundamental requirement that all other individuals be treated with respect for their humanity, which includes respecting the basic rights of everyone. Not employing violence against them is a minimum element of respecting them. In domestic life an individual may be intentionally harmed physically in a particular way only if a good justification applies specifically to that particular individual, for example, that she has by her own behaviour forfeited her right not to be harmed in that way. The same is true in most of international life.
In war in particular, by contrast, the rules permit those who qualify as combatants on one side to harm those who qualify as combatants on the other side without knowing anything about them as individuals, including whether they have done anything to forfeit their rights to physical security. In combat you attempt to kill unknown strangers before they kill you, although they do not know you as an individual any more than you know them as individuals. This is unavoidable once one has entered the circumstances of combat because those circumstances of kill-or-be-killed do not allow for the acquisition and thoughtful processing of the detailed information about particular individuals necessary for individual judgements about moral liability. This is why any attempt, such as David Rodin's and Jeff McMahan's, to base the rules for the conduct of war ultimately on individual moral liability is unhelpful.Footnote 11
The fundamental difficulty is that mortal combat is not the place for careful moral considerations about, and discriminating moral distinctions among, threatening adversaries, because sufficient quantities of the relevant information about differences among individual adversaries is not, and clearly cannot be, accessible in the time available for decisions. Even in what might be considered the best possible war, international law unavoidably permits much mutual killing and wounding of individual humans by other individual humans who know nothing important about each other except that they are directly participating in the military effort by the opposing side.Footnote 12 The specific rules for war are extremely different in content from the specific rules for ordinary domestic situations because they are adapted to radically different circumstances, especially including radically different quantities and qualities of information about individuals.
Any law that we seriously intend for people to abide by while they are attempting to kill before being killed must, then, be relatively simple and rely exclusively on readily observable features of the situation. The laws of war ought to be simple as well as generally exceptionless; exceptionlessness obviously contributes to simplicity. The primary, and evidently unchangeable, reason is the information poverty in the circumstances of combat: a fighter does not have the details about the motives, intentions, reasons, and past behaviour of individual adversaries that are needed to make the one-by-one judgements of moral liability called for by the theories of philosophers like McMahan and Rodin.
And a wide array of additional reasons why rules need to be simple and based on readily observable features could be assembled besides the information poverty. For example, first, during combat soldiers often become completely exhausted – war does not keep regular hours – and are lucky if they can see straight, much less reason subtly. Second, beloved comrades die in agony or receive fearful wounds, making calm moral assessment of perceived enemies virtually impossible psychologically. Third, until exhaustion takes over and actions become mechanical, everyone is edgy because deeply afraid of what will happen to them.Footnote 13 Combat is not a circumstance for circumspection and fine judgement. I do not belabour the point, however, because I have never known anyone acquainted with war to deny this.Footnote 14
The thickness of the fog of war must not, however, be exaggerated. Combatants on one side are not generally able to discriminate among different degrees of moral liability to being killed among different combatants on the other side, but all combatants must be able to discriminate between combatants and civilians. And the latter discrimination is usually possible, while the former is usually impossible, precisely because the difference between a civilian and a combatant is, and the difference between a morally liable combatant and a morally non-liable combatant is not, readily observable in the present from external features.Footnote 15
That required discriminations must be kept as simple as possible is widely acknowledged except by some analytic philosophers. Historian Geoffrey Best has observed: ‘What is oversimplified may be a better humanitarian working-tool than what has become endlessly complicated . . . The fuzzier the line and the further the spread of small print around it, the more opportunities must it offer to the risk-taker and the limit-stretcher.’Footnote 16 And General A. P. V. Rogers suggests: ‘If there is to be any hope that the law will be complied with, the rules must be as simple and straightforward as possible.’Footnote 17
This was why, for example, the 1907 Hague Regulations required a person who wished to qualify for the legal status of a combatant ‘to have a fixed distinctive emblem recognizable at a distance [and] to carry arms openly’.Footnote 18 The 1977 Geneva Protocol I eliminated the first of these two requirements in recognition of the widespread employment of guerrilla and other irregular tactics, but continued to insist upon signs readily visible during combat:
In order to promote the protection of the civilian population from the effects of hostilities, combatants are obliged to distinguish themselves from the civilian population while they are engaged in an attack or in a military operation preparatory to an attack . . . He shall retain his status as a combatant, provided that, in such situations, he carries his arms openly: (a) during each military engagement; and (b) during such time as he is visible to the adversary while he is engaged in a military deployment preceding the launching of an attack in which he is to participate.Footnote 19
During combat one must act upon what is evident in the present—reflective inquiries into the past are impossible. Similarly, one must act upon behaviour and appearance – consideration of current motive and present intention is limited to the degree that they are obvious from current behaviour and appearance, making mistakes inevitable. As Michael Herr, who carried a journalist's credentials, put it in contemplating what a North Vietnamese night assault on Khe Sanh would have been like: ‘Because if it came, it would be in a bloodswarm of killing, and credentials would not be examined.’Footnote 20 In recent years philosophers have produced extraordinarily subtle and sophisticated accounts of which aspects of one's motives, intentions, and past behaviour might be part of what is relevant to whether one was morally liable, on the basis of one's own individual moral features, to being killed, but as these theorists have come to acknowledge, the epistemological burdens of the analyses that would be required simply cannot be borne in the midst of mortal combat.Footnote 21 The epistemic demands of any adequate account of moral liability individual by individual are far too great to be satisfied during combat, even if fighters were in a fit physical and psychological state to assess morally relevant information. Therefore, we can conclude negatively but importantly that whatever requirements international laws for the conduct of war impose on who may be targeted by military force, they can neither turn upon nor reflect the moral liability, or any other deep characteristics, of particular individuals. Indeed, this is the main explanation of the negative side of our conclusion below that the content of the rules must be radically different from the content of any other rules, even though they must honour as many of our fundamental moral commitments as they can. Rules that require discriminations that depend on information that is not available cannot be applied.
So, on the one hand, at most the rules can limit such violent activities in relatively general ways. The goal of IHL can only be ‘to restrain the destructive force of war, while recognizing its inexorable necessities’.Footnote 22 Yet, on the other hand, if the violence cannot be limited sufficiently, the rules for war can never be morally justified because they depart too far from the fundamental moral principles that ought to guide all human action in all arenas and therefore fail to satisfy even minimally necessary moral requirements.
Many related differences between the rules for war and the rules for peaceful domestic life dictated by the differences in the circumstances to which they apply are equally obvious. The first criterion for IHL, then, is centred around an irresolvable tension: if the rules for war are to be morally justified, they must honour as many of our fundamental moral commitments as they can, but in light of the circumstances of organized lethal violence between groups in desperate conflict, the content of any relevant specific rules must be radically different from the content of the specific rules for any other circumstances.Footnote 23
2.2. The laws of war must be the same for all
While many lawyers take it as obvious that the laws of war must be the same for all fighters, a number of philosophers have recently challenged what has widely come to be called the ‘symmetry thesis’.Footnote 24 So I shall present diverse but converging reasons for it.
This is neither a new problem nor a problem that international lawyers have not wrestled with. Those who have formulated the international laws of war have been well aware of the fact that the rules they articulate, including the legal permissions they grant, will often be exploited by parties who are conducting wars that they ought not to be conducting at all – wars to which they ought not to have resorted. For example, in the Preamble to the most important recent multilateral treaty on the conduct of war, the 1977 Geneva Protocol I, the penultimate and ultimate paragraphs directly note this dilemma:
The High Contracting Parties . . . [4] expressing their conviction that nothing in the Protocol or in the Geneva Conventions of 12 August 1949 can be construed as legitimizing or authorizing any act of aggression or any other use of force inconsistent with the Charter of the United Nations, [5] reaffirming further that the provisions of the Geneva Conventions of 12 August 1949 and of this Protocol must be fully applied in all circumstances to all persons who are protected by those instruments, without any adverse distinction based on the nature or origin of the armed conflict or on the causes espoused by or attributed to the Parties to the conflict, have agreed . . . (original emphases).
Paragraph 4 takes the UN Charter to be the legal standard for justified resort to war and insists that nothing about the rules for the conduct of war to be specified grants legitimacy or authority to any party who is violating that standard, while paragraph 5 insists that the rules to be specified apply to all parties irrespective of the causes for which they fight, including, by clear implication, parties whose resort to war violates the Charter standard for the resort to war. One of the two leading commentaries on Protocol I observes that this wording ‘reflects the antagonism between on the one hand, the United Nations Law, which prohibits war and the use of force, and, on the other hand, humanitarian law, which establishes rules for situations which according to United Nations Law should not exist’.Footnote 25
The paradoxical-seeming strategy adopted by the international law for the conduct of war has been characterized by David Rodin as ‘the “symmetry thesis” which states that the same jus in bello [conduct] rights and obligations are held by combatants on both sides of any conflict’.Footnote 26 Combatants on both sides do in fact end up with the same legal rights and legal obligations, but one can see a part of the moral justification for the manner in which international law proceeds here if one analyses this outcome into two aspects.
2.2.1. Meaning (and moral limit on permissiveness)
The first aspect is that the correct rules for the conduct of war, both the legal rules and the moral rules, are the rules that should be followed by fighters whose resort to the use of violence is morally justified.Footnote 27 There is no need to tailor, nor any justification for tailoring, the content of the rules to suit parties who ought not to be fighting at all. There is no reason to make life easy legally for people who are acting immorally. The content of the laws ought to be formulated with only justified parties in mind: here is how you may fight if you are justified in fighting at all. If, as Protocol I assumes, a party is justified in employing force only if it is defending itself against prior attack, then the rules ought to be for parties who are defending themselves against attack. Parties should not be allowed to do anything that is not a necessary part of defending oneself.Footnote 28 This keeps legally permissible departures from morality to the minimum necessary for self-defence.
Therefore, the rules serve a function that can be described as ‘even if, only if’: even if your initial resort to war is justified, your continuing use of force is justified only if you abide by the rules for the conduct of war. The rules allow some actions that are necessary for parties who are justified in resorting to war. The rules do not have a suppressed premise saying, ‘since it makes no difference whether your resort to war is justified or not, you – whoever you are – may act as follows . . .’. On the contrary, if there is a suppressed premise, it is: ‘if you are justified in resorting to war, you may act as follows . . .’. This is the first aspect of the matter: the rules of conduct are a second requirement to be met after the first requirement of justified resort to force is met.
It is important to bear this first aspect in mind when one interprets Rodin's accurate statement that ‘the same jus in bello [conduct] rights and obligations are held by combatants on both sides of any conflict’. For, the second aspect of the matter is that in the international arena as it currently operates the rules will invariably be invoked by both those who are morally and legally justified in a resort to force and those who are not. We have no international institutional arrangements to guarantee that those who do not satisfy the principles for the resort to war will not avail themselves nevertheless of the rules for the conduct of war.
2.2.2. International political context
Now, the fact that the laws for the conduct of war may be invoked by both sides is not a result of some inadequacy in the content of those laws, but a result of the fact that in the existing international arena, states claim as an aspect of their sovereignty the prerogative of deciding for themselves when war may be initiated.Footnote 29 So Rodin's statement is for now true in fact, but only because of two kinds of failings, the first of which may not be remediable, but the second of which may be subject to improvement. These failings are not failings in IHL itself, but failings in the national agents that international law must govern.
The first failing is cognitive: people come to believe that they are justified in going to war when they are not. I am not saying that this mistake is excusable (or not)—simply that it regularly occurs.Footnote 30 This point about cognition is a purely empirical observation. One could explore many aspects of this moral error: when is it the result of ignorance? When is the ignorance culpable? And so on. I leave all those important issues aside here, and simply note that such misguided war is a frequent occurrence. The result is that in fact parties who are not justified in resorting to war do so and then help themselves to the international laws for the conduct of war including all the legal permissions granted in those laws. In the terms of the Protocol commentary, parties act in violation of Charter law and thereby unfortunately bring themselves into the domain of humanitarian law.
The second failing is in international political institutions. We currently have no legitimate political institution with the authority to declare definitively in advance, in the case of every party, whether that party's resort to war is morally justified.Footnote 31 This second failing, then, is that at present we have no effective institutional remedy for the first failing, the cognitive moral errors about resorting to war.Footnote 32 If there were some kind of international court that could rule authoritatively in advance on each party's resort to war, inform them whether they were allowed to go to war, and rule authoritatively on whether they qualified to exercise the privileges granted by the laws of war, parties without a justification for war could be forbidden by the institution in question from availing themselves of the laws for the conduct of war because of having failed to satisfy the standards for the resort to war. But this would of course amount to forbidding them to go to war at all. If some court had the universal legitimacy to be able to make such rulings authoritatively beforehand, they would be in a position to stop wars by ruling that at least one of the parties had no right to proceed. If a state were to acknowledge the authority of such an institution, it would be surrendering an aspect of its sovereignty, as currently understood.Footnote 33 The simple fact, once again, is that we now have no such institution with the power and authority to stop all parties from going to war when they are unjustified. States continue to assert a prerogative ultimately to decide whether they are justified in going to war.
The UN Charter prohibits the resort to war except as a defensive response to an attacker who has already violated this very Charter prohibition. The Achilles heel of the system for now is that the interpretation of when a resort to military force is defensive has so far been successfully asserted by states to be part of their sovereignty. The UN prohibits aggression, but states effectively decide, at least in the first instance, when they are aggressors.
This establishes, then, another simple but important negative point: it is not some fault in the formulation of IHL that is responsible for its being in practice applicable to parties to wars that have no justification for having gone to war. The faults are, first, cognitive – people, including powerful political elites, get important matters wrong – and, second, institutional – we do not have widely accepted and therefore powerful political institutions that can put matters right when cognitive errors are committed by political elites. If we are to have any laws at all for the conduct of war, they are in fact going to be invoked by all parties, which, as Rodin puts it, means by ‘combatants on both sides of any conflict’. What is crucial is this: this cannot be changed by changing the content of the laws, because the fact of appeal to the laws of conduct by both sides is not a result of some peculiarity of the law's content. That would be a mistaken empirical diagnosis that would lead us to focus on the wrong problem.Footnote 34
Furthermore, one ought in any case to be glad if, once some party has made the wrong decision about whether to resort to war, it at least chooses to abide by IHL. Every act of violence this party commits will be morally wrong, but if it abides by IHL, it may commit fewer wrongs than it would do if it fought its unjustified war in an unrestrained fashion. Restrained unjustified warfare is a somewhat lesser evil than unrestrained unjustified warfare, especially for those who would perish in the latter, but not in the former. IHL can restrain war; it cannot ensure that it is justified.
2.3. Response to Rodin
David Rodin has mounted a powerful response to my earlier presentations of these arguments. Even if one is willing to ‘accept for the moment’ that ‘the criteria of individual liability to harm are impossible to apply to the circumstances of war’, Rodin correctly notes that any attempt ‘to assess the moral acceptability of practices that would require divergent moral codes’ must be conducted against the background of two critical points: first, ‘we start with a defeasible presumption against such deviation’, and, second, any justification of such a deviation ‘can only be achieved by reference to the values and principles of ordinary morality itself’.Footnote 35 Because, as I observed above, morality is all of a piece, Rodin is right both that any rules that deviate from ordinary rules certainly bear the burden of proof, and that this very deviation must itself be justified on the basis of ordinary values, not some exotic alternative values that we have no reason suddenly to embrace. The argument must take the general form of showing that the rights, interests, or values that are normally best served by the usual rules are in some special circumstances, such as war, better served by unusual rules. And Rodin proceeds to offer a troubling proposed analogy to what I have argued above concerning war in the form of a nineteenth-century opponent of the abolition of slavery who argues instead for humanitarian regulation and claims that ‘if we are to regulate the practice of slavery we must recognise that the content of those regulations must be very different from the content of ordinary morality and must allow the owning of other persons’.Footnote 36
As a relatively minor preliminary, I am not an opponent of the abolition of war who thinks regulation is preferable. Few of the wars during my lifetime have actually been justified, and I believe that almost all of them ought to have been prevented. My primary concern is that the fact that almost all wars ought not to be fought is all the more reason why, if they cannot in fact be prevented, they ought to be limited. Secondarily, I also believe that occasionally military force could be used successfully and proportionally to stop an evil so extreme – for example, the Rwandan genocide of 1994 – that even the evils of the war that would be entailed are lesser ones.
However, much more important is the fact that while what Rodin says about proposals for ‘divergent moral codes’ is correct, what I am defending here is not a divergent moral code but an international legal code that, while it is indeed divergent in its content from the content of morality, is morally justified in the role that it plays: ‘The goal is deeply morally informed international laws’.Footnote 37 How can a set of legal rules for an unusual situation whose content does not replicate the content of the moral rules for ordinary circumstances be better justified morally by reference to the very same morality that grounds those moral rules than a set of rules for the same unusual situation that did replicate the content of the ordinary moral rules would be? Clearly the answer must be along the lines of an argument that there is a morally important role to be played in the unusual situation that can be better played by non-standard rules than by standard ones. What could this role be?
The short answer is: the role of limiting the violence of each of two conflicting forces who disagree about which of the two is justified in employing violence. The fact that the two sides disagree about which of them is justified in employing violence means that any rules that are to be respected by both sides must not presuppose which of the two sides is justified. If the rules presuppose that one of the two sides is not justified, that side will ignore the rules as not in fact applicable to it (since it believes itself to be justified and the rules presuppose that it is not). In that case only one of the two sides would be constrained by the rules. This is why the rules for the conduct of war, such as the First Geneva Protocol, are neutral concerning which of the two sides is in violation of the United Nations Charter. The Protocol provides the same rules for the conduct of the violator and the conduct of the defender of the Charter.
Given that this is understandable, how can it be, Rodin asks, morally justified? The forces of the violator are granted legal permission to kill and wound the forces of the defender – surely, Rodin suggests, much is amiss morally. My response, unfortunately, is: yes and no – much is amiss, but not with the rules for the conduct of war. What is morally amiss is that one side is fighting without moral justification – one side is violating the moral and legal principles for the resort to war. What Rodin condemns as the ‘symmetry’ of the rules for the conduct of war (the jus in bello) is made possible by what he calls the ‘independence’ of the rules of conduct from the rules of resort. But the alternative to the independence of rules for conduct from rules of resort would appear to be rules of conduct that somehow presuppose which of the adversaries is justified in its resort to war.
Rodin clearly thinks that at this point I begin to bark up the wrong tree: ‘those [like Rodin himself] who favour an asymmetric interpretation of jus in bello do not propose rules of war that are non-neutral in the sense that they have the form “if you are a just combatant do x”; “if you are an unjust combatant do y”.’Footnote 38 In fact, he concedes the only point that I am insisting on: ‘the in bello rules themselves are perfectly neutral and address just and unjust combatants in precisely the same way’. This is what I think matters, and I am simply adding the further claim that such neutral legal rules are morally justified in their role of the laws guiding the conduct of war because – this is an empirical hypothesis – rules of this kind – that is, laws neutral between the two sides, or symmetrical in Rodin's usage – do as much to limit unjustified killing and wounding in war as can be done if they also maximize the protection of civilians while permitting the mutual killing and wounding of combatants.
Are the combatants on the unjustified side fully morally justified, then, as long as they abide by the neutral laws for the conduct of war? Absolutely not: full moral justification has two necessary conditions, compliance with the best rules for the conduct of war and compliance with the best principles for the resort to war. Those on the unjustified side have failed to comply with the principles of resort even if they have complied with the rules of conduct. The crucial distinction is between the moral justification for neutral laws of the conduct of war themselves and the moral justification for those who abide by those laws. Those who abide by morally justified laws of the conduct of war are not themselves fully morally justified unless they have also complied with the principles for the resort to war. The laws can be fully morally justified laws – my thesis here – even when those who comply with them are not fully justified agents because agents at war must also comply with a second necessary set of principles, those for resort.
My remaining disagreement with Rodin is, I think, of less general theoretical importance. His next sentence immediately following the last one quoted above says: ‘It is simply that unjust combatants [combatants in violation of the principles of resort] cannot be in compliance with them [the rules of conduct] (as the rules specifying lawful force in self-defence cannot be satisfied by an aggressor).’ On the contrary, I believe that a combatant who is in violation of the principles of resort can still comply with the rules of conduct. Rodin and I disagree about whether compliance with laws of conduct is possible because of underlying disagreements about such matters as the interpretation of proportionality, but space is not available here to pursue those issues.Footnote 39
3. The possible trade-off: stringency and the price of compliance
Insisting on the same rules for all creates opportunities, unfortunately, for dubious trade-offs. Other things equal – and many other things would need to be equal – the more stringent the content of a set of rules, the greater the temptation on the part of self-interested parties to flaunt the rules. Any restrictions whatsoever on the conduct of war are going to help to motivate some temptation to ignore the rules imposing the restrictions. So there is a possible trade-off between stringency and likely breadth of compliance; that is, numbers and types of warring parties who will actually comply with the rules.Footnote 40 More permissive rules would, all else equal, probably garner wider compliance. And, to move to one extreme case, laws for the conduct of war that more or less guaranteed the defeat of any party who followed them would be at least highly likely, if not certain, to be ignored. At this very abstract level we can see one possible trade-off for the moral justifiability of IHL as follows: the more permissive the laws of war are, the greater the range of warring parties that are likely to be willing to comply with them, but the more permissive the laws of war are, the farther they depart from our fundamental moral commitments and thus the more difficult it becomes to justify their content morally.
Once again, this is exactly how international lawyers have long understood the problem. The standard introduction to IHL begins as follows: the content of the laws for the conduct of war is the outcome of a tension between military necessity and ‘humanitarianism’.Footnote 41 ‘Military necessity’ stands, roughly, for what it takes to attain military success – to ‘win’, if you like, although succeeding need not mean victory in any straightforward sense.Footnote 42 ‘Humanitarianism’ consists of the fundamental moral commitments. The standard idea of the tension is this. If, on the one hand, IHL made no significant concession to what it takes to achieve success militarily, parties at war would simply have ignored them; the laws of war would have been consigned to the realm of idealistic scribbling. If, on the other hand, IHL made no significant room for fundamental moral commitments, IHL would be pointless since, after all, its purpose is to restrain fighters from doing whatever they think it takes to succeed militarily.Footnote 43 The most general form of the tension, then, is this: if ‘humanitarianism’ does not get its due, the whole exercise of having laws for the conduct of war is useless from the start, but if ‘military necessity’ does not get its due, the laws will be futile in the end, because no one who wants to avoid defeat will pay any attention to them. Fundamental moral commitments cannot be fully honoured because military necessity minimally requires actions that violate the rules (moral and legal) for peaceful domestic life – it requires extensively ‘killing people and breaking things’. This is what makes it war, at least as wars are now fought.Footnote 44 But if our fundamental moral commitments really are our fundamental commitments, we can justify indulging their violation only so much and only for very good reason.
Now the preceding paragraph is quite general. In order to sharpen the issues, we need to be slightly more precise about the almost hopelessly vague and contentious matter of military necessity.Footnote 45 What I have called the concessions to military necessity must minimally take the form of leaving it possible for some parties to have some reasonable prospect of military success. It is vital, however, to see that there is no need whatsoever to formulate IHL either so that every party to a conflict has a prospect of military success – obviously impossible, since at least one party must lose – or so that any kind of party has a guarantee of military success. Instead, IHL need do only this: leave some kind of fighting chance for some kinds of parties in some kinds of wars.Footnote 46 If IHL left a fighting chance for no one, it would be ignored. So, one way to view the issues is to ask: (i) how much of a fighting chance for (ii) which kinds of parties to (iii) which kinds of wars? What fighting chance for whom is it necessary for the laws to allow in order to gain wide compliance with the laws, other things equal?
I do not mean to suggest that the only means of gaining compliance is by making laws more permissive. Hope of reciprocity from compliance, fear of reputational losses from non-compliance, sanctions, and prosecutions under international criminal law are among the other factors that may not be equal, and we can choose to use them rather than concessions on stringency.Footnote 47 But consider for now the narrow question: what would be the price in permissiveness of the laws for various degrees of compliance, other things equal? This, on the one hand, is a question of fact, political and psychological: overly stringent laws can be self-defeating. But overly permissive laws will be morally unjustifiable. The normative question that would then arise is: is the degree of permissiveness necessary to gain a certain degree of compliance, whatever in fact it is, morally justifiable? That, on the other hand, is a matter of right and wrong. This is the tension. Nothing guarantees in advance that the moral price in permissiveness of compliance will not be too high – that what would work politically and psychologically will not be wrong morally. If the moral price in permissiveness of a workable IHL is too high, we ought to reject the trade-off; and we will need to turn to means other than concessions for increasing compliance with existing law, such as tougher sanctions, more aggressive enforcement, or more energetic shaming.
4. Limits of compromise: terrorism and anti-terrorism
Today we face a choice that is a clear example of the potential trade-off between stringency and compliance. From the beginning of modern international law in the late nineteenth century, what was above referred to as ‘humanitarianism’ has consisted above all in the protection of civilians.Footnote 48 The prohibition on intentional targeting of civilians, the principle of distinction, is exceptionless – described by the International Court of Justice in 1996 as ‘intransgressible’.Footnote 49 Protection against foreseeable harm to civilians is less strict – how permissive depends on how loosely the complementary principle of proportionality is interpreted.
Compliance with distinction and with proportionality impose very different prices on different kinds of fighters. For terrorists, on the one hand, compliance with distinction would put them out of operation because their primary tactic is the intentional killing of civilians.Footnote 50 Obviously, then, terrorists cannot comply with the principle of distinction as it now stands (and has long stood) and remain terrorists. If we wanted to make it easier for them to comply with the ‘basic rule’ of IHL, we would need to weaken the rule. It certainly could not remain exceptionless and would presumably need to leave the number of civilians intentionally killed to some degree to the discretion of the terrorists. I will not fuss over exactly how to formulate a hypothetical principle of distinction-for-terrorists that would accommodate their main tactic. Let us quickly say, merely for illustrative purposes, that one principle that could be acceptable to them might be: never target any more civilians than necessary in order to have a fighting chance of military success.
On the other hand, the evil twin of the terrorist is the kind of counterterrorist who believes that the way to discourage terrorism is to deter it by inflicting disproportionate casualties on populations from which terrorists come. Thus, this counter-terrorist is more interested in seeing the principle of proportionality compromised than in seeing the principle of discrimination compromised (although that may also be thought to help). Again I will not fuss over exactly how to state a hypothetical appropriately weakened principle of proportionality-for-counterterrorists, but it might say something roughly like this: never foreseeably kill more civilians than necessary in order to deter terrorist attacks.
If international law were ever to compromise the principle of distinction to suit terrorists and to compromise the principle of proportionality to suit counter-terrorists, the compliance of the terrorists and the counterterrorists with the new weakened versions of each principle might be gained. But this ‘compromise’ would, I believe, give away the moral game entirely. At least in my loose illustrative formulations, each principle would in effect say: use your own judgement about how many civilians you kill.Footnote 51 This was roughly the rule used by the RAF and the USAF in the Second World War and by the USAF in the Korean War, and it produced total war. So I think it is quite evident that this moral price for gaining compliance – mutual escalation at the cost of civilians – is too high. We can honour our fundamental moral commitments much more fully by instead stringently enforcing IHL as it exists.
Consider, on the contrary, an argument given by Michael Gross for more permissive rules concerning civilians, where the specific rules he is advocating are not the crude, extreme ones I have used above for illustrative purposes, but are considerably more moderate. I want here, however, to concentrate on the grounds for the changes, not the specific changes proposed, because the grounds tie back into the general tension between stringency and compliance. Gross's normative premise is a putative ‘right to a fighting chance’,Footnote 52 combined with a number of substantial empirical claims about the nature of fighting between guerrilla organizations and anti-guerrilla forces. Gross attributes the idea of this right to Charles Chaumont, whom he quotes as proposing the following:
In order to remain objective and credible, humanitarian law must allow every party an equal chance in combat. If a norm of this body of law is incompatible with this principle and makes it impossible from the outset for one of the parties to have any prospect of victory, it is better not to draft such a norm at all.Footnote 53
To his credit Gross immediately notes that this view is ‘astonishing’ and ‘extraordinary’, but he takes it seriously nevertheless and indeed explicitly characterizes it as centring on a right.
This putative right to a fighting chance serves as a normative premise for an argument for greater permissiveness about attacks on civilians, when combined with various empirical assumptions about the nature of such fighting. The basic theses are that if guerrillas are to enjoy the supposed right to a fighting chance, they may need to kill civilians through terrorist tactics; and if the counterterrorists are to enjoy a right to a fighting chance, they too similarly need to attack more civilians, namely the so-called ‘associated civilians’ who contribute indirectly to the effectiveness of the guerrilla forces using terrorism. So each side would be allowed to escalate its attacks to some degree in order to have more of a fighting chance, which here becomes a supposedly greater chance of winning for each through escalation at the cost of civilians by both.
It is necessary to emphasize immediately that the assertion that there is a right to a fighting chance is indeed an extraordinarily strong claim. It needs to be, in effect, a claim about a natural right because it was originally advanced by Chaumont as a condition on the acceptability of international laws, thus logically preceding international laws and constituting a necessary condition for their acceptability. I would not deny that there are some natural rights, but establishing any particular one requires commensurately powerful arguments. Gross argues as follows:
Oppressed groups as well as nations fighting the tyranny of a rogue state deserve a reasonable fighting chance when diplomacy fails. There are two reasons to support a right to a fighting chance. First, a fighting chance is integral to the idea of just cause. It makes little sense to acknowledge one group's right to fight oppression or another's right to intervene in the affairs of a repressive regime and then use the law of armed conflict to deny either group the means to do so. Second, and perhaps more practically, any state or nonstate party denied a fighting chance by the conventions of war has little reason whatsoever to support humanitarian norms of conduct.Footnote 54
I shall return to Gross's first reason below, but I begin with his second reason, which is presumably his impetus for proposing to allow mutual escalation in attacks against civilians and which directly raises the issue of the moral costs of compliance. As I noted above, we in fact have several means of attempting to gain compliance with IHL besides making it more permissive, so Gross is incorrect in claiming that ‘any state or nonstate party denied a fighting chance by the conventions of war has little reason whatsoever to support humanitarian norms of conduct’. We can give them other reasons in the form of prosecution or other sanctions. Nevertheless, weaker constraints are presumably easier to abide by, and people are, other things equal, more likely to obey rules that demand less of them, so greater permissiveness might seem to be one route to take.
The two fundamental purposes of IHL are to respect fundamental moral commitments while allowing some kinds of military action to go forward with some kind of prospect of success. The latter certainly need not, however, be because any groups are believed to have a right to a fighting chance, but might be in part because more groups would ignore the laws of war if they had no prospect of military success because of the stringency of the rules. The rules for war cannot, as I have argued above, be incompatible with all military success by every party, or they will be generally ignored and be completely ineffectual. It is perfectly true that there would be no reason to grant any kind of legal combatant privilege at all if not to make it possible for some kinds of groups to have some kind of prospect of military success. But the greater permissiveness advocated on the basis of Gross's second reason includes much more in two respects: the reasoning extends the permissiveness beyond ‘oppressed groups as well as nations fighting the tyranny of a rogue state’ to an unlimited range of groups, and the demand is for an equal chance for each group with whoever is their opponent, which seems to mean an unlimited range of opponents.
The supposed right to an equal chance of military success irrespective of who is the chosen opponent is totally implausible. Why should the international community switch into the business of equalizing various groups’ chances of succeeding in warfare when it has since 1945 been in the business of discouraging warfare as much as possible? Does every aggressor have a right to a chance of victory equal to the chance that its victim will be able to defend itself successfully? I cannot imagine any reason why it would. Is every group that attacks a superpower entitled to a chance of victory equal to the superpower's chances? That would require arrangements that seem preposterous to me.Footnote 55
The other question is which kinds of groups or military forces would supposedly be entitled to the right to a fighting chance. What is the identity of the bearers of such a right? Who is meant to be ‘any party’? Surely not every group who can afford to buy AK-47s and RPGs thereby acquires a right to a fighting chance against whomever they happen to want to attack! Such a strong natural right could belong only to groups who satisfied a number of fairly stiff criteria concerning who they are and whom they are going to fight. We would need some good reasons why some particular group is naturally entitled to a fighting chance to win a war against some particular other group.Footnote 56
Yet the answer that seems to flow from Gross's second reason is that it is whichever groups threaten otherwise not to abide by existing international law: we buy their compliance by weakening our demands in response to their threats. But this genuinely is a slippery slope, and we need to have instead some principled basis for choosing the groups (which Gross's first reason – not yet discussed – attempts to provide). And the principled basis must focus very sharply on respecting fundamental moral commitments, given the types of military action we feel compelled to tolerate. I see no reason at all to tolerate the current tactics of either terrorists or counterterrorists, who seem to be locked in a destructive symbiosis that is threatening to what few constraints there now are on war. It is, however, the burden of whoever proposes weakening the protections for civilians to give good reasons why whichever groups are insisting on the new permissiveness should have their tactics indulged beyond what IHL now allows rather than prosecuted as war crimes.
Before we turn to Gross's first reason, it is also worth observing that even if one grants that reciprocity is as important as he assumes throughout his book, the conclusion that there should be mutual escalation in attacks on civilians by no means follows. Demanding that both sides respect the existing immunity of civilians is as fully reciprocal as allowing both sides to attack more civilians. The argument from reciprocity does not lead directly to mutual escalation. Mutual restraint can embody at least as much reciprocity.Footnote 57 Appeals to reciprocity do nothing in themselves to advance the specific thesis that the rules ought to be more permissive.
Gross's first reason is more plausible: ‘a fighting chance is integral to the idea of a just cause’. The implications of this first reason could obviously not be more different from the evident implications of the second, for now only fighters with a just cause, rather than every group which can get its hands on enough weapons to be threatening, qualify for the putative right to the fighting chance of military success. No counterterrorists or terrorists whose causes are unjust would qualify. Nevertheless, this reason too encounters insuperable difficulties.
The notion that a fighting chance is integral to the idea of a just cause seems to mean something like: if a group has a just cause, the group has a right to use whatever military means would be necessary to give them a fighting chance for military success. First, this flatly denies what is widely acknowledged to be the fundamental principle of the laws of war: ‘In any armed conflict, the right of the Parties to the conflict to choose methods or means of warfare is not unlimited.’Footnote 58 However just one's cause, there are means one may not employ – otherwise, there is no limiting war, since all sides will in fact do whatever the just are allowed to do, by believing or pretending to believe that their cause is just. Second, some groups with just causes are in such a weak position that they could not possibly win a military victory without using nuclear weapons – this may well include some current terrorist groups. Is it then objectionable for the laws of armed conflict to deny them the means they would need in order to succeed? Hardly.
The fatal problem here comes from attempting to build military success into just cause. People who have a just cause ought to succeed, of course – this is what it means to say they have a just cause. But to claim that they ought to succeed militarily is to make a much more specific and completely groundless claim. Many just causes simply cannot be advanced by military means, and many groups with just causes are, sadly perhaps, in absolutely no position to advance their cause through warfare – they need to work in other ways.Footnote 59 It would indeed be hypocritical to claim that one agreed that some group had a just cause and then pass laws that deprived them of all possible means of succeeding in advancing their cause. But it need not be hypocritical to deprive them of some military means, e.g. terror bombing, in order to protect the basic rights of other people.
Third, even if Gross's first reason were sound, the suggestion would be unworkable for the usual reasons, already discussed above. In order to make the rules more permissive for the groups who qualified by having a just cause, the international arena would need some legitimate institution, widely accepted cross-culturally, that could declare authoritatively which groups were the groups with a just cause; the groups specified by the institution would then be allowed to use the more permissive rules and attack more civilians. To agree to respect such an institution is to agree not to initiate war on one's own – if parties could do the first, they could do the second and would then not need the institution. So there is no feasible way to implement a suggestion that justified sides could have privileges that unjustified sides could not have.
Now Gross's argument is only one, and others are being given for encroaching on the rights now guaranteed to civilians. In principle, each argument must be confronted on its own terms, but a general dilemma emerges for advocates of loosening IHL. Any suggestion that the rules be made more permissive only for some parties – presumably the justified – and not for others faces the same kind of unworkability faced by all asymmetric rules and discussed more fully earlier. But any suggestion that the rules be loosened for all parties whatsoever faces the daunting task of demonstrating some feature of all parties to contemporary conflict that is a good reason why they all should be allowed to engage in forms of violence beyond all those already allowed by current IHL, forms of violence that will violate the basic moral rights of yet additional categories of persons. At the moment the excessive violence both of the terrorists and of the counterterrorists seems to be producing a downward spiral that I can see no good reason to encourage with legal immunity. As Nietzsche observed, ‘Whoever fights monsters should see to it that in the process he does not become a monster.’Footnote 60 Even a clear-eyed and hard-headed view of IHL requires no concessions to terrorists or anti-terrorists.
This illustrates the more general point that one can acknowledge important military and political realities faced by IHL without conceding a moral minimum. Whether the particular moral minimum thereby preserved is high enough to allow the overall moral justifiability of IHL is, of course, a much larger issue.
